REAL-TIME ANALYTICS ON IOT DEVICES WITH CLOUD SUPPORT

Abstract

The rapid growth of the Internet of Things (IoT) has created an ecosystem where billions of connected devices generate massive quantities of facts in real time. Efficient processing of this data is critical for applications such as healthcare observing, smart cities, industrial automation, and intelligent transportation systems. Traditional analytics frameworks often struggle to handle the high velocity, variety, and volume of IoT data, necessitating the integration of cloud computing platforms that provide scalable storage and computational resources [1]. This paper presents an in-depth study of real-time analytics on IoT devices supported by cloud infrastructures. A hybrid architecture is proposed, combining lightweight edge processing on IoT nodes with scalable cloud services for advanced analytics and visualization.

We evaluate the proposed system using a case study in healthcare monitoring, where wearable IoT devices track patient vitals such as heart rate and oxygen saturation. Mathematical models are introduced to quantify end-to-end latency, bandwidth requirements, and energy consumption. Experimental results demonstrate that edge-assisted cloud analytics reduce latency by 32% and optimize bandwidth utilization by 27% compared to cloud-only processing. Furthermore, anomaly detection models, such as z-score and ARIMA-based forecasting, are employed to identify irregular patient conditions with an accuracy of 95%. The findings highlight the potential of cloud-assisted IoT analytics to achieve scalable, reliable, and energy-efficient real-time decision-making.